def __test_txt(self, image_file, expected_box_file, lang='eng'):
image_file = "tests/data/" + image_file
expected_box_file = "tests/tesseract/" + expected_box_file
with codecs.open(expected_box_file, 'r', encoding='utf-8') \
as file_descriptor:
expected_boxes = self.builder.read_file(file_descriptor)
expected_boxes.sort()
boxes = tesseract.image_to_string(Image.open(image_file),
lang=lang,
builder=self.builder)
boxes.sort()
self.assertTrue(len(boxes) > 0)
self.assertEqual(len(boxes), len(expected_boxes))
for i in range(0, min(len(boxes), len(expected_boxes))):
try:
# python 2.7
self.assertEqual(type(expected_boxes[i].content), unicode)
self.assertEqual(type(boxes[i].content), unicode)
except NameError:
# python 3
self.assertEqual(type(expected_boxes[i].content), str)
self.assertEqual(type(boxes[i].content), str)
self.assertEqual(boxes[i], expected_boxes[i])
def __test_txt(self, image_file, expected_box_file, lang='eng'):
image_file = "tests/data/" + image_file
expected_box_file = "tests/tesseract/" + expected_box_file
with codecs.open(expected_box_file, 'r', encoding='utf-8') \
as file_descriptor:
expected_boxes = self.builder.read_file(file_descriptor)
expected_boxes.sort()
boxes = tesseract.image_to_string(Image.open(image_file), lang=lang,
builder=self.builder)
boxes.sort()
self.assertTrue(len(boxes) > 0)
self.assertEqual(len(boxes), len(expected_boxes))
for i in range(0, min(len(boxes), len(expected_boxes))):
try:
# python 2.7
self.assertEqual(type(expected_boxes[i].content), unicode)
self.assertEqual(type(boxes[i].content), unicode)
except NameError:
# python 3
self.assertEqual(type(expected_boxes[i].content), str)
self.assertEqual(type(boxes[i].content), str)
self.assertEqual(boxes[i], expected_boxes[i])
def extractText(pPath):
lImage = Image.open(pPath)
lResult = tesseract.image_to_string(lImage, lang="deu")
print (lResult)
if (lResult != None):
return 1, pPath
else:
return 0, pPath
def recognition():
from os import chdir, environ
from tesseract import image_to_string
from Image import open
path = environ.get("HOME")
im = open("blob.jpg")
text = image_to_string(im)
chdir(path + "/alpr/latest/")
return text
def get_authcode(self,url):
self.headers['Host'] = 'authcode.jd.com'
self.headers['Referer'] = 'https://passport.jd.com/uc/login'
response = self.session.get(url, headers = self.headers)
with open('authcode.jpg','wb') as f:
f.write(response.content)
# authcode = input("plz enter authcode:")
authcode = image_to_string('authcode.jpg', False)
return authcode
def recognition():
from os import chdir,environ
from tesseract import image_to_string
from Image import open
path = environ.get("HOME")
im = open("blob.jpg")
text = image_to_string(im)
chdir(path+"/alpr/latest/")
return text
def loadFile():
ftypes = [('Supported Image files', '*.png'), ('All files', '*')]
fl = tkFileDialog.askopenfile(filetypes=ftypes, title='Choose a file')
if fl != None:
content = image_to_string(Image.open(fl))
'''Create a Text widget'''
txt = Text()
txt.pack(fill=BOTH, expand=1)
txt.insert(END, content)
loadImage(fl)
def im2string(name):
#file_name='LITE_peg.jpeg'
file_name = '2' + name
text = tesseract.image_to_string(file_name, True, ' -psm 7 digits')
#im = Image.open(file_name)
#text = pytesseract.image_to_string(im,config='outputbase digits')
print(text)
return text
def post(self):
self.set_header("Content-Type", "text/plain")
self.write("You sent a file with name " +
self.request.files.items()[0][1][0]['filename'])
# make a "memory file" using StringIO, open with PIL and send to tesseract for OCR
self.write(
image_to_string(
Image.open(
StringIO.StringIO(
self.request.files.items()[0][1][0]['body']))))
def get_authcode(self, url):
'''
验证码
'''
self.headers['Host'] = 'authcode.jd.com'
self.headers['Referer'] = 'https://passport.jd.com/uc/login'
response = self.session.get(url, headers=self.headers)
with open('authcode.jpg', 'wb') as f:
f.write(response.content)
authcode = image_to_string('authcode.jpg', False)
return authcode
def tostring(self):
#2 fois plus grand en bicubique (il me semble que cela donne de meilleur résultat)
self.img = self.img.resize((self.img.size[0] * 2,
self.img.size[1] * 2), Image.BICUBIC)
# convertit en noir et blanc (niveaux de gris)
self.img = self.img.convert("L")
if self.invert:
self.invertVid()
if self.bold:
self.tobold()
self.img = self.img.resize((self.img.size[0] * 2,
self.img.size[1] * 2), Image.NEAREST)
return image_to_string(self.img)
def __test_txt(self, image_file, expected_output_file, lang="eng"):
image_file = "tests/data/" + image_file
expected_output_file = "tests/tesseract/" + expected_output_file
expected_output = ""
with codecs.open(expected_output_file, "r", encoding="utf-8") as file_descriptor:
for line in file_descriptor:
expected_output += line
expected_output = expected_output.strip()
output = tesseract.image_to_string(Image.open(image_file), lang=lang)
self.assertEqual(output, expected_output)
def __test_txt(self, image_file, expected_output_file, lang='eng'):
image_file = "tests/data/" + image_file
expected_output_file = "tests/tesseract/" + expected_output_file
expected_output = ""
with codecs.open(expected_output_file, 'r', encoding='utf-8') \
as file_descriptor:
for line in file_descriptor:
expected_output += line
expected_output = expected_output.strip()
output = tesseract.image_to_string(Image.open(image_file), lang=lang)
self.assertEqual(output, expected_output)
def __test_txt(self, image_file, expected_box_file, lang="eng"):
image_file = "tests/data/" + image_file
expected_box_file = "tests/tesseract/" + expected_box_file
with codecs.open(expected_box_file, "r", encoding="utf-8") as file_descriptor:
expected_boxes = self.builder.read_file(file_descriptor)
expected_boxes.sort()
boxes = tesseract.image_to_string(Image.open(image_file), lang=lang, builder=self.builder)
boxes.sort()
self.assertEqual(len(boxes), len(expected_boxes))
for i in range(0, min(len(boxes), len(expected_boxes))):
self.assertEqual(boxes[i], expected_boxes[i])
def __test_txt(self, image_file, expected_box_file, lang='eng'):
image_file = "tests/data/" + image_file
expected_box_file = "tests/tesseract/" + expected_box_file
with codecs.open(expected_box_file, 'r', encoding='utf-8') \
as file_descriptor:
expected_boxes = self.builder.read_file(file_descriptor)
expected_boxes.sort()
boxes = tesseract.image_to_string(Image.open(image_file), lang=lang,
builder=self.builder)
boxes.sort()
self.assertEqual(len(boxes), len(expected_boxes))
for i in range(0, min(len(boxes), len(expected_boxes))):
self.assertEqual(boxes[i], expected_boxes[i])
def post(self):
self.set_header("Content-Type", "text/html")
self.write("<html><body>")
self.write("You sent a file with name " + self.request.files.items()[0][1][0]['filename'] +"<br/>" )
# create a unique ID file
tempname = str(uuid.uuid4()) + ".jpg"
myimg = Image.open(StringIO.StringIO(self.request.files.items()[0][1][0]['body']))
myfilename = os.path.join(os.path.dirname(__file__),"static",tempname);
# save image to file as JPEG
myimg.save(myfilename)
self.write("<img src=\"static/" + tempname + "\" /><br/>")
# do OCR, print result
self.write(image_to_string(myimg))
self.write("</body></html>")
def extract_pdf_text(src_pdf_file_path):
os.environ[
"TESSDATA_PREFIX"] = '/opt/local/share' # this is required otherwise tesseract complains about file permissions
with open(src_pdf_file_path, 'rb') as src_pdf_file:
pdf_reader = PyPDF2.PdfFileReader(src_pdf_file)
# pdfReader.numPages
# 19
for page_index in range(pdf_reader.numPages):
page = pdf_reader.getPage(page_index)
image = pdf_page_to_png(page, resolution=72)
# extract_pdf_page_images(page)
tmp_img_path = '/tmp/titi.png'
cv2.imwrite(tmp_img_path, image)
text = tesseract.image_to_string(
Image.open(tmp_img_path)) # , lang='deu')
print(text)
def __test_txt(self, image_file, expected_box_file, lang='eng'):
image_file = "tests/data/" + image_file
expected_box_file = "tests/tesseract/" + expected_box_file
boxes = tesseract.image_to_string(Image.open(image_file), lang=lang,
builder=self.builder)
boxes.sort()
with codecs.open(expected_box_file, 'r', encoding='utf-8') \
as file_descriptor:
expected_boxes = self.builder.read_file(file_descriptor)
expected_boxes.sort()
self.assertEqual(len(boxes), len(expected_boxes))
for i in range(0, min(len(boxes), len(expected_boxes))):
for j in range(0, len(boxes[i].word_boxes)):
self.assertEqual(type(boxes[i].word_boxes[j]),
type(expected_boxes[i].word_boxes[j]))
self.assertEqual(boxes[i], expected_boxes[i])
def test_write_read(self):
original_boxes = tesseract.image_to_string(Image.open("tests/data/test.png"), builder=self.builder)
self.assertTrue(len(original_boxes) > 0)
(file_descriptor, tmp_path) = tempfile.mkstemp()
try:
# we must open the file with codecs.open() for utf-8 support
os.close(file_descriptor)
with codecs.open(tmp_path, "w", encoding="utf-8") as file_descriptor:
self.builder.write_file(file_descriptor, original_boxes)
with codecs.open(tmp_path, "r", encoding="utf-8") as file_descriptor:
new_boxes = self.builder.read_file(file_descriptor)
self.assertEqual(len(new_boxes), len(original_boxes))
for i in range(0, len(original_boxes)):
self.assertEqual(new_boxes[i], original_boxes[i])
finally:
os.remove(tmp_path)
def main():
q = 0
k = True
j = 0
while k:
#k = False
#image_org = grab_screen(region=(0,30,1250,750))
image_org = cv2.imread(file_paths[q])
j = j + 0.5
lower = np.array([55, 100, 100])
upper = np.array([90, 255, 255])
#image_org = cv2.bilateralFilter(image_org,9,75,75)
speed_img = image_org[610:700, 1090:1200]
image_org = cv2.medianBlur(image_org, 5)
image = cv2.cvtColor(image_org, cv2.COLOR_BGR2HSV)
#speed_img =
print(image_to_string(speed_img))
#image_org = cv2.GaussianBlur(image_org, (3, 3), 0)
#back = cv2.imread(bd)
#back = cv2.resize(back, (image.shape[0], image.shape[1]))
#crop_img = img[y:y+h, x:x+w]
#colour_extract = image_org[500:750, 200:950]
#image_org = cv2.GaussianBlur(image_org, (5, 5), 0)
#colour_extract = cv2.cvtColor(colour_extract, cv2.COLOR_BGR2HSV)
#lower, upper = colour_extracting(colour_extract)
#print(colour_extract.shape)
#print(lower, upper)
#image = cv2.cvtColor(image_org, cv2.COLOR_BGR2HSV)
#mask = cv2.inRange(image_org, lower, upper)
#image = cv2.bitwise_and(image_org, image_org, mask= mask)
#linesP = cv2.HoughLinesP(cv2.cvtColor(image, cv2.COLOR_BGR2GRAY), 1, np.pi / 180, 50, None, 50, 10)
#if linesP is not None:
# for i in range(0, len(linesP)):
# l = linesP[i][0]
# cv2.line(image, (l[0], l[1]), (l[2], l[3]), (0,0,255), 3, cv2.LINE_AA)
cv2.imshow("window", speed_img)
#cv2.imshow("crop",colour_extract)
#cv2.imshow("org", image_org)
if cv2.waitKey(25) & 0xFF == ord('q'):
cv2.destroyAllWindows()
break
def test_write_read(self):
original_boxes = tesseract.image_to_string(
Image.open("tests/data/test.png"), builder=self.builder)
self.assertTrue(len(original_boxes) > 0)
(file_descriptor, tmp_path) = tempfile.mkstemp()
try:
# we must open the file with codecs.open() for utf-8 support
os.close(file_descriptor)
with codecs.open(tmp_path, 'w', encoding='utf-8') as file_descriptor:
self.builder.write_file(file_descriptor, original_boxes)
with codecs.open(tmp_path, 'r', encoding='utf-8') as file_descriptor:
new_boxes = self.builder.read_file(file_descriptor)
self.assertEqual(len(new_boxes), len(original_boxes))
for i in range(0, len(original_boxes)):
self.assertEqual(new_boxes[i], original_boxes[i])
finally:
os.remove(tmp_path)
def post(self):
self.set_header("Content-Type", "text/html")
self.write("<html><body>")
self.write("You sent a file with name " +
self.request.files.items()[0][1][0]['filename'] + "<br/>")
# create a unique ID file
tempname = str(uuid.uuid4()) + ".jpg"
myimg = Image.open(
StringIO.StringIO(self.request.files.items()[0][1][0]['body']))
myfilename = os.path.join(os.path.dirname(__file__), "static",
tempname)
# save image to file as JPEG
myimg.save(myfilename)
self.write("<img src=\"static/" + tempname + "\" /><br/>")
# do OCR, print result
self.write(image_to_string(myimg))
self.write("</body></html>")
def process_arlista(href):
img = Image.open(StringIO.StringIO(download(href)))
img = img.point(threshold)
img = img.convert("1")
csikok = arlista_darabol(img)
arlista = []
datum = date.strftime(date.fromtimestamp(int(re.search("\d{10}",href).group(0))), "%Y-%m-%d")
for csik in csikok[1:]:
sor = [datum]
for kocka in csik:
nagy = kocka.resize((kocka.size[0]*2, kocka.size[1]*2), Image.NEAREST)
string = tesseract.image_to_string(nagy, lang='csapi')
if csik.index(kocka) is 0 and string is "":
break
if csik.index(kocka) > 0: # ha nem az áru megnevezése...
scrubbed = scrub(string)
sor = sor + list(scrubbed)
if csik.index(kocka) is 0:
sor.append(string)
if len(sor) > 1 and sor[2]:
arlista.append(sor)
csviro = csv.writer(open('arlista.csv', 'a'), quoting=csv.QUOTE_NONE)
csviro.writerows(arlista)
def ocr(filename):
f = open(os.path.join(UPLOAD_FOLDER, filename))
return image_to_string(Image.open(f))
def post(self):
self.set_header("Content-Type", "text/plain")
self.write("You sent a file with name " + self.request.files.items()[0][1][0]['filename'] )
# make a "memory file" using StringIO, open with PIL and send to tesseract for OCR
self.write(image_to_string(Image.open(StringIO.StringIO(self.request.files.items()[0][1][0]['body']))))
def ocr(filename):
f = open(os.path.join(UPLOAD_FOLDER, filename))
return image_to_string(Image.open(f))
def to_text(self, im, boxes=False, improve=True, lang='por'):
if improve:
im = self.improve_image(im)
if boxes:
lang=None
return tesseract.image_to_string(im, lang=lang, boxes=boxes).decode('utf-8')
pixdata = img.load()
# Make the letters bolder for easier recognition
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][0] < 90:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][1] < 136:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][2] > 0:
pixdata[x, y] = (255, 255, 255, 255)
img.save("input-black.gifIF")
# Make the image bigger (needed for OCR)
im_orig = Image.open('input-black.gifig = im_orig.resize((1000, 500), Image.NEAREST)')
ext = ".tif"
big.save("input-NEARESTxt")
# Perform OCR using tesseract-ocr library
from tesseract import image_to_string
image = Image.open(input-NEAREST.tifrint image_to_string(image))
import Image
from tesseract import image_to_string
print(image_to_string(Image.open('E:\\aa.png')))
print(image_to_string(Image.open('E:\\aa.png'), lang='eng'))
import numpy as np
import argparse
import cv2
import tesseract
t2 = cv2.imread('cir2.png')
thresh = cv2.cvtColor(t2, cv2.COLOR_BGR2GRAY)
thresh = cv2.adaptiveThreshold(thresh, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
cv2.THRESH_BINARY_INV, 91, 1)
print(tesseract.image_to_string(thresh))
# count, labels, stats, centroids = cv2.connectedComponentsWithStats(thresh)
# for i in range(1,count):
# t2 = cv2.circle(t2, (int(centroids[i,0]), int(centroids[i,1])), 5, (0, 255, 0, 0), 5)
cv2.imshow('circles', thresh)
cv2.imshow('centers', t2)
cv2.waitKey()
def GetImgValue(url, HostPage, UserAgent, cookies, split=None):
headers = {}
headers['Accept'] = 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'
headers['Connection'] = 'keep-alive'
headers['Host'] = url.split('/')[2]
headers['Referer'] = HostPage
headers['User-Agent'] = UserAgent
imgData = requests.get(url, headers=headers, cookies=cookies).content
im = cStringIO.StringIO(imgData)
img = Image.open(im)
img = img.convert("RGBA")
pixdata = img.load()
enh = ImageEnhance.Contrast(img)
enh.enhance(1.3).show("30% more contrast")
# Make the letters bolder for easier recognition
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][0] < 90:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][1] < 136:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][2] > 0:
pixdata[x, y] = (255, 255, 255, 255)
if split == 'LR' or split == 'TB':
img1 = Crop(img=img, split=split, half=1)
img2 = Crop(img=img, split=split, half=2)
img1.save("input-black1.gif", "GIF")
img2.save("input-black2.gif", "GIF")
# Make the image 1 bigger (needed for OCR)
im_orig = Image.open('input-black1.gif')
big1 = im_orig.resize((1000, 500), Image.NEAREST)
ext = ".tif"
big1.save("input-NEAREST1" + ext, "TIFF")
image1 = Image.open('input-NEAREST1.tif')
# Make the image 2 bigger (needed for OCR)
im_orig = Image.open('input-black2.gif')
big2 = im_orig.resize((1000, 500), Image.NEAREST)
ext = ".tif"
big2.save("input-NEAREST2" + ext, "TIFF")
image2 = Image.open('input-NEAREST2.tif')
recaptcha1 = image_to_string(image1)
recaptcha2 = image_to_string(image2)
recaptcha = recaptcha1 + " " + recaptcha2
else:
if split == 'TOP':
img = Crop(img=img, split=split)
img.save("input-black.gif", "GIF")
# Make the image bigger (needed for OCR)
im_orig = Image.open('input-black.gif')
big = im_orig.resize((1000, 500), Image.NEAREST)
ext = ".tif"
big.save("input-NEAREST" + ext, "TIFF")
image = Image.open('input-NEAREST.tif')
recaptcha = image_to_string(image)
return recaptcha
import Image
from tesseract import image_to_string
pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"
print image_to_string(Image.open('scan2.png'))
def parse_txt(img_path, lang=None):
# TODO: validate img_path
return image_to_string(Image.open(img_path), lang=lang)
'''import cv2
import numpy as np
import pyzbar.pyzbar as pyzbar
image = cv2.imread("bca.jpg")
decodedObjects = pyzbar.decode(image)
for obj in decodedObjects:
print("obj", obj)
print("Type:", obj.type)
print("Data: ", obj.data, "\n")
cv2.imshow("Frame", image)
cv2.waitKey(0)
'''
from pillow import Image
from tesseract import image_to_string
print( image_to_string(Image.open('IMG_20200116_151537__01.jpg')) )
print( image_to_string(Image.open('IMG_20200116_151537__01.jpg'), lang='eng') )
def find(image_path):
print(image_to_string(Image.open(image_path)))
print(image_to_string(Image.open(image_path), lang='eng'))
# Make the letters bolder for easier recognition
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][0] < 90:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][1] < 136:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][2] > 0:
pixdata[x, y] = (255, 255, 255, 255)
img.save("input-black.gifIF")
# Make the image bigger (needed for OCR)
im_orig = Image.open('input-black.gif')
big = im_orig.resize((1000, 500), Image.NEAREST)
ext = ".tif"
big.save("input-NEAREST" + ext)
# Perform OCR using tesseract-ocr library
from tesseract import image_to_string
image = Image.open('input-NEAREST.tif')
print(image_to_string(image))
from PIL import Image
from tesseract import image_to_string
print(image_to_string(Image.open('pic.jpg')))
print(image_to_string(Image.open('test-english.jpg'), lang='eng'))
import Image
from tesseract import image_to_string
print image_to_string(Image.open('test.png'))
print image_to_string(Image.open('test-english.jpg'), lang='eng')
from PIL import Image
from tesseract import image_to_string
print(image_to_string(Image.open("114.jpg")))
import pytesseract as pyt
import tesseract as tst
from PIL import Image
img = Image.open('edited.jpg')
#print(image_to_string(Image.open('edited.jpg')))
print(tst.image_to_string(img))
pixdata[x, y] = (255, 255, 255, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][0] < 90:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][1] < 136:
pixdata[x, y] = (0, 0, 0, 255)
for y in xrange(img.size[1]):
for x in xrange(img.size[0]):
if pixdata[x, y][2] > 0:
pixdata[x, y] = (255, 255, 255, 255)
img.save("input-black.gif", "GIF")
# Make the image bigger (needed for OCR)
im_orig = Image.open('input-black.gif')
big = im_orig.resize((1000, 500), Image.NEAREST)
ext = ".tif"
big.save("input-NEAREST" + ext)
# Perform OCR using tesseract-ocr library
from tesseract import image_to_string
image = Image.open('input-NEAREST.tif')
print image_to_string(image)
